Forge FateAn ETF research design exercise: questions, assets, and comparison rules
Defines an ETF research question, asset universe, and comparison rules, introducing the scope of the subsequent synthetic exercises.
Quant Research Design and Synthetic Exercises
Eight parts on research questions, acquisition criteria, and timing, with synthetic accounting and record checks. Parts 7–8 include a runnable public example. Historical ETF baseline performance is covered separately in the earlier series; this is not a completed ML or live-trading system.
- 1.
Quant & Data Research - 2.
Quant & Data Research Before acquiring ETF data: sources, price definitions, and quality checks
Designs source, usage, price-definition, and quality checks before acquisition; a preparation record, not a completed data collection report.
- 3.
Quant & Data Research Why shift(1) Is Not Enough: A Time Contract for Signals, Execution, and Labels
Separate signal, execution, return, and label timing. Use synthetic unit tests to check information leakage and preprocessing boundaries.
- 4.
Quant & Data Research Start a Backtest with Positions, Not Returns: One-Period Accounting for Buy and Hold vs. Momentum
Check holdings, cash, and ending values for buy-and-hold and momentum using three synthetic assets, establishing accounting rules before market testing.
- 5.
Quant & Data Research Would Adding Transaction Costs Change the Conclusion? Checking Accounting and Risk Paths in a Synthetic Portfolio
Use a synthetic portfolio to check transaction-cost accounting and distinguish what ending returns, turnover, and maximum drawdown reveal.
- 6.
Quant & Data Research Designing a logistic-regression comparison: timing, evaluation, and failure handling
Designs timing, splits, preprocessing, and failure rules for comparing logistic regression with a baseline; model training and empirical evaluation are not included.
- 7.
Quant & Data Research Building a synthetic experiment recorder: success, failure, and a runnable example
Runs public code that records success and failure for synthetic probability inputs; not a market prediction or trained-model result.
- 8.
Quant & Data Research Rechecking synthetic experiment records: a public example and design-series review
Checks reruns and record comparisons for a public synthetic example, separating these checks from ETF performance and ML validation.